The ultimate goal of this Component is to identify mouse genes which contribute to the Down syndrome-like phenotype in mouse mutants trisomeric for regions syntenic with human chromosome 21. The success of the Ts65Dn model, following on that of the trisomy 16 model, clearly demonstrates the utility of mouse models of Down syndrome. At the same time, the genotype/phenotype study of human partial trisomies have proven to be a robust mapping method, although the rarity of human partial trisomies has made progress slow. We propose to apply the partial trisomy mapping method to the mouse mode, but rather than rely on rare, and unsystematic, natural trisomies, we will use the Cre/lox site specific recombination system to engineer a specific, systematic set of segmental deletions and duplications, either "adding" or "subtracting" large regions from the Ts65Dn trisomy model. Specifically, we will introduce lox sites by homologous recombination in murine embryonic stem cells at various loci in three regions: on chromosome 16 within 1 Mb of the AML1 locus and within a 2 Mb regions between the Wv and Erg loci, and on chromosome 17 in a 1 Mb regions region around Cbs. We will then catalyze the recombination between these lox sites, either in Es cell in culture or in fertilized oocytes, by transient expression of the Cre recombinase. Through analysis of mutant mice carrying these duplication or deletions in the Ts65Dn background, we will correlate trisomy of specific regions with particular aspects of the DS phenotype; and then candidate genes from these regions, identified in the Gene Identification Core, will be tested directly.